MPRIS media keys continue affecting Firefox while media.hardwaremediakeys.enabled is false
Categories
(Core :: Audio/Video, defect)
Tracking
()
People
(Reporter: lexlexlex, Unassigned)
References
Details
Attachments
(1 file)
|
57.04 KB,
text/plain
|
Details |
User Agent: Mozilla/5.0 (X11; Linux x86_64; rv:96.0) Gecko/20100101 Firefox/96.0
Steps to reproduce:
- Have a play/pause media key assigned which works with DBus MPRIS.
- Open at least 1 audio player that's not Firefox which supports DBus MPRIS global media keys.
- Ensure
media.hardwaremediakeys.enabledis set tofalse(boolean value) in Firefox'about:configmenu. - Have at least 1 tab open with a YouTube video.
- Actuate the play/pause media key which was verified to work in step 1.
Actual results:
The background media player and Firefox both toggled play/pause.
Expected results:
Only the background media player was expected to toggle play/pause.
Comment 1•4 years ago
|
||
Thanks for the report! Would you mind navigating to your about:support page, copying the the raw data in a file, and then uploading it here?
I tried to reproduce this by opening a chromium browser alongside with Firefox but I cannot reproduce this, unfortunately. Do you still see the virtual media control buttons in the notification center once you set media.hardwaremediakeys.enabled to false?
Thank you for your response and your attempt to reproduce!
I've attached about:support data above. I don't know what you mean by "notification center". I don't see any "virtual media control buttons" anywhere. I'm using MPRIS via a play/pause media keystroke from a hardware device.
The two clients which are simultaneously affected in my case are QMMP 2.0.2 and Firefox Developer Edition 96.0b10. I'm running Manjaro with XFCE.
I'm glad to help figure this out further.
I think this can be tested without a hardware media play/pause key by having both Firefox and any other MPRIS player running, then entering playerctl -a play-pause in a terminal. My theory is that if handling media keys is disabled, then Firefox should not respond to MPRIS messages via playerctl.
Does doing that reproduce the issue for you? It's possible that's not a correct way to reproduce it though, in case Firefox really strictly polls for hardware media keys directly instead of using media.hardwaremediakeys.enabled for MPRIS properly.
Comment 5•4 years ago
|
||
I couldn't reproduce this issue on my Ubuntu 20.04 as well. So on the step4 mentioned in comment0, when you disable media.hardwaremediakeys.enabled and is playing youtube on Firefox, does playerctl -l show that Firefox is on that list?
Also, would you mind to capture the debug log by using MOZ_LOG=MediaControl:5 while doing the steps mentioned in comment0?
Thank you.
Updated•4 years ago
|
Aha! Thanks for the playerctl -l tip! That helped me find the real cause of this, and it seems it's the Plasma Browser Integration extension. Here's the output from that command from before and after disabling that extension:
[alex@alex-pc ~]$ playerctl -l
plasma-browser-integration
qmmp
[alex@alex-pc ~]$ playerctl -l
qmmp
[alex@alex-pc ~]$
Disabling it made the issue go away. So, I guess a browser extension can override the state of media.hardwaremediakeys.enabled and/or provide its own MPRIS listener. I had figured that since I'm using XFCE and not Plasma, the "Plasma Browser Integration" plugin would be irrelevant to anything, but I was wrong.
This seems invalid to me now. Thanks for your help.
Description
•